Dataset for the paper: Managing Systems Evolving in Space and Time: Four Challenges for Maintenance, Evolution and Composition of Variants

The dataset contains ground-truth variants and their configurations for all scenarios described in the paper: Managing Systems Evolving in Space and Time: Four Challenges for Maintenance, Evolution and Composition of Variants. The ground-truth variants for feature location and composition of variants with features is in the folder named by <systemName>_Variants_With_Features.rar The ground-truth for feature revision location and composition of variants with feature revisions is in the folder named by <systemName>_Variants_With_Feature_Revisions.rar The additional scenario is in the folder named by LibSSH_Variants_With_Feature_Revisions_ExtraScenario.rar For every scenario use as input variants the variants inside a folder named by inputConfigurations. For every scenario use as ground-truth variants of new configurations the variants inside a folder named by newConfigurations. The scenarios for feature location are split according to the number of variants, therefore the first scenario, for example, has the first variant of the file named by Inputconfigurations.csv. The second scenario must have the first variant + the second variant of the file named by Inputconfigurations.csv. The third scenario must have the first three variants of the file named by Inputconfigurations.csv, and so on, according to the Table 1 in the paper. Regarding composing new configurations with features, use the new configurations in the file Newconfigurations.csv or NewConfigurations.txt and as ground-truth variants the variants in the folder newConfigurations. The scenarios for feature revision location are split according to the number of Git commits, therefore the first scenario, for example, has all the variants of the first Git commit in the file named by inputConfigurations.csv. The second scenario must have all the variants from the first 5 Git commits in the file named by inputConfigurations.csv, and so on, according to the Table 2 in the paper. Note: For every line of the file inputConfigurations.csv, the first column represents the number of a Git commit, the second column contains the hash of a Git commit and the following columns contain the feature revisions of a configuration. Regarding composing new configurations with feature revisions, use the new configurations in the file newConfigurations.csv and as ground-truth variants the variants in the folder newConfigurations.本数据集包含论文《时空演化系统管理：变体维护、演化与组合的四大挑战》（Managing Systems Evolving in Space and Time: Four Challenges for Maintenance, Evolution and Composition of Variants）中所述所有场景的基准变体（ground-truth variants）及其配置。面向特征（feature）定位与带特征的变体组合的基准变体存储于名为`<systemName>_Variants_With_Features.rar`的压缩文件夹中；面向特征修订（feature revision）定位与带特征修订的变体组合的基准真值存储于名为`<systemName>_Variants_With_Feature_Revisions.rar`的压缩文件夹中。额外场景存储于名为`LibSSH_Variants_With_Feature_Revisions_ExtraScenario.rar`的压缩文件夹中。对于所有场景，输入变体均取自名为`inputConfigurations`的文件夹内的变体文件；新配置的基准变体均取自名为`newConfigurations`的文件夹内的变体文件。特征定位类场景按照变体数量进行划分：例如第一个场景对应`Inputconfigurations.csv`文件中的首个变体，第二个场景对应该文件中的前两个变体，第三个场景对应前三个变体，其余场景依此类推，具体划分规则可参考论文中的表1。在执行带特征的新配置组合任务时，需使用`Newconfigurations.csv`或`NewConfigurations.txt`文件中的新配置，且基准变体取自`newConfigurations`文件夹内的变体文件。特征修订定位类场景按照Git提交次数进行划分：例如第一个场景对应`inputConfigurations.csv`文件中首次Git提交对应的所有变体，第二个场景对应该文件中前5次Git提交对应的所有变体，其余场景依此类推，具体划分规则可参考论文中的表2。注意：`inputConfigurations.csv`文件的每一行中，第一列为Git提交序号，第二列为Git提交哈希值，后续各列为某一配置的特征修订项。在执行带特征修订的新配置组合任务时，需使用`newConfigurations.csv`文件中的新配置，且基准变体取自`newConfigurations`文件夹内的变体文件。